Interesting observation, but the variability is established during the front end of the flight. Also, notice the term "simulation". Garbage in = garbage out. Slugs just don't come with steering wheels to put them back on track.
OneMinute, Here's something funny; That video was actually produced by Litz. Apparently it's not uncommon that folks misinterpret it. And in the LongRangeHunting thread, Litz provides the following clarification;
Quote
This video is often misunderstood. Note that the illustration on the left is not showing the path of the bullet, it's showing the angle that the nose is pointed. the view on the right is showing the bullet path from the shooters point of view. You can see how tiny the actual 'corkscrew' path of the bullet is.
This video is the result of modeling which means it's only as accurate as the inputs, and there are a lot of inputs needed to make a 6-DOF model work. I'm sure that there are some imperfections in some of the inputs. However, when you consider that the model would have to be wrong by more than 100 times in order to show the level of group convergence that's commonly claimed (1 MOA at 100 and 1/2 MOA at 200), it becomes far fetched to accept epicyclic swerve as the explanation.
Here's the 2.0 version of Litz' video that might make clearer what exactly it represents;
